Look Inside Kona Electric Battery And See Real Capacity

So, given that CDC, CCC and cell voltages are measurements and expected to be reliable we can deduce that the SoC derivation from that data needed correction. Have you checked the SoC and corresponding voltage against the LG Chem curve, referenced in blue on my graph above?

 

A final graph just for the record showing the contribution of regen. Net depleted coulombs in blue = consumed in red - regened in green.
Where the lines loop to the left is where I gain SoC on downhills.

 

I did check in the past the voltage vs SoC and it is scatter like yours. I will check against le LG Chem curve you presented. How can I get the data points of that curve?
This curve shows that in the last 5 % SoC, the voltage is dropping drastically. Given that the Coulomb are constant over the entire SoC range, this would mean there is a lot less energy per % SoC available in this last 5% compare to the first 95%. I never discharge my battery under 8%, so no data to confirm this.
Does this curve reflects the usable range of the Kona battery pack? There is talks that Hyundai keeps a buffer zone at both ends.

 

Here is a voltage vs display SoC for a 100% to 11%:

 

Here's an image of the data I entered from the E63 cell datasheet. I don't think I can attach an .xls here, sorry, you'll have to type it in.

 

I discuss this in a previous post. I recorded a discharge from 100% SoC display to 11%. I plotted the Net energy used versus display SoC. Since the curve is not perfect linear, I did a 2nd order polynomial regression and obtain this result:




Using the polynomial equation, I create a table were I calculate the energy used for each % SoC as the battery discharges toward 0%. There are to much scatters in recorded data to obtain a linear result. The regression equation gives a much better result. Here is a snippet of this table:



Using the table, I calculated the delta energy between each %SoC to see many kWh in theory are used for each % SoC from 100 to 0 % Here is a snipped of it:



This table produced the following graph:



I don't have any data below 8% SoC so this part is pure theorical. This result shows that there is about 34 kWh available from 100% to 50% and 30 kWh left in the last 50% SoC.

 

I replot my previous battery voltage vs SoC graph. I divided the battery pack voltages by 98 to give the average cell voltages instead. I added the LG OVC points on top of it.

 

That's even a better fit than I had. Perhaps I need to take a trip with less hills.

 

Yesterday, I was able to trigger a new SoC correction again. To acheave it, I start the car and turn off all accessories (climat control, radio, lights,...). I leave the car running as it for about 30 minutes. My SoC yesterday was at 39.5% and it suddently shift to 42.5%. The SoC where CCC = CDC was at 65.5% since July 2nd, with this SoC correction, it has moved to 68.5%. I drove about 3500Km since the last correction.

This seems to be a recalibration of the SoC based on some BMS variable. Since most of people will rarely leave their Kona turned On not moving for an extended period of time, this would mean that their SoC will be off calibration at some point?...

It would be interesting if someone else could try this.